Pigasus Awards & Sheldrake

[juryjone]

Yes, I agree that a survey about whether I've ever guessed who was calling me before I answered the phone is pretty darn meaningless. Is that what we're talking about?

~~ Paul

To me, the problem is that the claim is that the person is thought of before the telephone rings, but the experiment has the person identified after the phone starts ringing (am I getting that correct?). Also, when this happens "in the field", there is the potential of the next call being from an unknown, whereas in the experiment the next call is definitely from the known. I'm not seeing how this can be considered to be testing the claim.

However, I know that testing a claim and trying to test a hypothesis can be different. I haven't read the paper so I don't know if Sheldrake outlined his hypothesis more clearly than the quote I got from davidsmith73.

 

[davidsmith73]

How is that random?

If the output from an RNG is influenced by a Global Consciousness, then the output for this experiment cannot be random.

The RNG output is random with respect to emotional vs calm target selection, according to the null hypothesis. The timing of the target presentation is deterministic and set at time 0.

Personally, I don't prescribe to the idea of a "global consciousness" because I think observation of the output is necessary for PK. I think the GCP results, if due to PK, are down to the experimenters observing the output of the RNG's and biasing it according to their expectations.

If you're asking about how to tell whether the RNG's in the presentiment experiment are being affected by "global consciousness" or not, I don't know. You'd better ask Radin.

 

[davidsmith73]

To me, the problem is that the claim is that the person is thought of before the telephone rings, but the experiment has the person identified after the phone starts ringing (am I getting that correct?).

"Immediately" after the phone is picked up, according to the paper! Its not necessarily the case that subtle cues could be leaked in that situation but, surely, a simple protocol change whereby the choice is made before the phone rings would clear up that problem. Perhaps Sheldrake has, ahem, different standards of evidence. I don't understand it myself.

Also, when this happens "in the field", there is the potential of the next call being from an unknown, whereas in the experiment the next call is definitely from the known. I'm not seeing how this can be considered to be testing the claim.

I can see where you (and Zep) are coming from. But consider how difficult it would be to quantify a comparison with chance if you were to conduct an experiment where, for example, the recipient was trying to guess the name of random people from the phone book who were about to call, the names of which were unknown to the recipient. What if they got some names correct? How would you determine, quantitatively, the likelyhood that was just due to chance? Not to mention that it makes sense to try to increase the likelyhood of observing an effect in the lab. If examples "in the field" suggest that the effect is most common when it's known callers (yes, we know that normal explanations are on offer to explain why this may be the case), then it makes sense to do an experiment with known callers in order to establish an effect relatively quickly.

 

[CFLarsen]

The RNG output is random with respect to emotional vs calm target selection, according to the null hypothesis. The timing of the target presentation is deterministic and set at time 0.

Personally, I don't prescribe to the idea of a "global consciousness" because I think observation of the output is necessary for PK.

Why? A tree can't fall in the woods without someone watching?

I think the GCP results, if due to PK, are down to the experimenters observing the output of the RNG's and biasing it according to their expectations.

So, you think that it is possible to influence the RNG just by observing it?

If you're asking about how to tell whether the RNG's in the presentiment experiment are being affected by "global consciousness" or not, I don't know. You'd better ask Radin.

Radin doesn't answer my emails.

 

[davidsmith73]

Why? A tree can't fall in the woods without someone watching?

Its just a hypothesis I happen to like. It makes sense to me. If PK exists then I don't see how a coherent theory can be produced that doesn't involve observation as part of the process. What else could a PK mechanism involve if not a relationship with observation? I don't know.

So, you think that it is possible to influence the RNG just by observing it?

Observing some kind of meaningful output. Like you observe the output of a coin toss as a meaningfull relationship between "heads" vs "tails". If you want to influence "heads vs tails", you have to observe "heads vs tails" to know whether the process has been influenced. So it seems to me that observation is a necessary part of PK. If we take away observation then we don't really have any PK do we?

 

[davidsmith73]

Is there a telephone telepathy experiment somewhere that appears to be sound?

~~ Paul

The Lobach and Bierman study looks to be tighter than the Sheldrake study.

http://www.sheldrake.org/articlesnew/pdf/Lobach.pdf

Ersby's already mentioned this one. Its interesting that the hit rate is much lower than the Sheldrake study, although still significant. So the Sheldrake study may have an inflated hit rate due to methodological problems. Perhaps not surprising. Lets hope the Chris French collaboration comes to fruition.

 

[CFLarsen]

Its just a hypothesis I happen to like. It makes sense to me. If PK exists then I don't see how a coherent theory can be produced that doesn't involve observation as part of the process. What else could a PK mechanism involve if not a relationship with observation? I don't know.

Here we are again, back at the old believer's "I happen to like" credo.

It doesn't matter what you "happen to like". Either a phenomenon exists, or it doesn't.

Observing some kind of meaningful output. Like you observe the output of a coin toss as a meaningfull relationship between "heads" vs "tails". If you want to influence "heads vs tails", you have to observe "heads vs tails" to know whether the process has been influenced. So it seems to me that observation is a necessary part of PK. If we take away observation then we don't really have any PK do we?

Who is doing the observation? The one with the powers?

 

[davidsmith73]

Here we are again, back at the old believer's "I happen to like" credo.

It doesn't matter what you "happen to like". Either a phenomenon exists, or it doesn't.

Correct. I doesn't matter to the existence of the anomalous results, whether I like the hypothesis I mentioned or not.

Who is doing the observation?

The one or ones observing the output of the REG.

The one with the powers?

I don't know. Perhaps its the first person to observe the results. Perhaps subsequent people can affect the results in an additive way. Thats for experiments to find out.

 

[Paul C. Anagnostopoulos]

mmm. Occams razor?

Ooh, I wouldn't bring up Occam in a psi discussion.

If the output from an RNG is influenced by a Global Consciousness, then the output for this experiment cannot be random.

Shhh, you're not supposed to mention that. I bet you're the kind of guy who would bring up the control problems in prayer studies, too.

Interesting paper: http://m0134.fmg.uva.nl/publications/2002/expectationbias_PA2002.pdf

~~ Paul

 

[Paul C. Anagnostopoulos]

The Lobach and Bierman study looks to be tighter than the Sheldrake study.

http://www.sheldrake.org/articlesnew/pdf/Lobach.pdf

I find it odd that the paper mentions nothing about how they controlled for caller ID and ring patterns. Also, where's the raw data?

~~ Paul

 

[Paul C. Anagnostopoulos]

Here's a replication attempt that failed:

http://216.239.51.104/search?q=cach...telepathy"+protocol&hl=en&ct=clnk&cd=14&gl=us

~~ Paul

 

[Zep]

"Immediately" after the phone is picked up, according to the paper! Its not necessarily the case that subtle cues could be leaked in that situation but, surely, a simple protocol change whereby the choice is made before the phone rings would clear up that problem. Perhaps Sheldrake has, ahem, different standards of evidence. I don't understand it myself.

Now you are catching on to the issues we have been raising for some time. These are fairly basic and obvious issues when it comes to test protocol design. They bear thinking about BEFORE rushing forward to the test itself and subsequent publication...of questionable results.

I can see where you (and Zep) are coming from. But consider how difficult it would be to quantify a comparison with chance if you were to conduct an experiment where, for example, the recipient was trying to guess the name of random people from the phone book who were about to call, the names of which were unknown to the recipient. What if they got some names correct? How would you determine, quantitatively, the likelyhood that was just due to chance? Not to mention that it makes sense to try to increase the likelyhood of observing an effect in the lab. If examples "in the field" suggest that the effect is most common when it's known callers (yes, we know that normal explanations are on offer to explain why this may be the case), then it makes sense to do an experiment with known callers in order to establish an effect relatively quickly.

Considering the original claims as provided, it is actually NOT unreasonable at all. And yes, it is certainly feasible to design a practical and representational protocol that uses a sample population of, say, 1000 random people from the phone book that are not reasonably known to the recipient. That allows quantification of the results as well.

The issue with the "limited set of known callers" protocol, while valid and interesting in itself, is that it is NOT testing the claim being made, even in a prototype mode. The original claim is a test of an actual "psi" capability - the ability to view/recognise another person at a distance without using normal means (in fact, the phone call is merely a synchronising mechanism, and not really vital in itself). Instead, the limited protocol tests something else (primarily the ability to beat the odds). Therefore any effect measured probably cannot be related to or translated to the original claim. Being good at the latter does not imply strength at the former.

 

[juryjone]

The issue with the "limited set of known callers" protocol, while valid and interesting in itself, is that it is NOT testing the claim being made, even in a prototype mode.

I certainly agree with that.

The original claim is a test of an actual "psi" capability - the ability to view/recognise another person at a distance without using normal means (in fact, the phone call is merely a synchronising mechanism, and not really vital in itself).

I don't get this part, however. Where does it say that the recipient of the call "sees' the caller in the action of calling them? That would imply precognition, seeing a future event. However the fact that it's called "telephone telepathy" means that they're looking for telepathy instead. Either the recipient transmits the desire for a phone call to the caller, or the caller transmits the intention to the recipient. In that case the actual phone call is crucial, and not a "synchronizing mechanism".

 

[CFLarsen]

The one or ones observing the output of the REG.

How will you ever falsify that theory?

I don't know. Perhaps its the first person to observe the results. Perhaps subsequent people can affect the results in an additive way. Thats for experiments to find out.

How will you find out? The moment the observers observed the data, it could change, right?

Shhh, you're not supposed to mention that. I bet you're the kind of guy who would bring up the control problems in prayer studies, too.

Why would you say that?

 

[Zep]

I don't get this part, however. Where does it say that the recipient of the call "sees' the caller in the action of calling them? That would imply precognition, seeing a future event. However the fact that it's called "telephone telepathy" means that they're looking for telepathy instead. Either the recipient transmits the desire for a phone call to the caller, or the caller transmits the intention to the recipient. In that case the actual phone call is crucial, and not a "synchronizing mechanism".

I was trying to describe the protocol of determining who is calling after the phone rings but before it is picked up. It's a test of clairvoyence.

In this case, the phone call simply synchronises the recipient's attempt to determine the remote caller; the caller performs the sync trigger action by calling. It could just as easily be a light-switch at one end and a light at the other, with the recipient trying to determine who flicked the switch.

The protocol where the recipient "wishes" or "expects" for a particular person to call, and then they do, is another different idea again! That's precognition as well as clairvoyence...and remarkably easy to fudge impressive results with.

 

[Professor Yaffle]

Here's a replication attempt that failed:

http://216.239.51.104/search?q=cach...telepathy"+protocol&hl=en&ct=clnk&cd=14&gl=us

~~ Paul

From the above study (towards the end):

Finally there remains the possibility that our result is due to an intrinsic property of Psi not to show upstable within the same experimental context. Not to find the same effect again in a replication is a well-known result within parapsychology and has also occurred in our work before (see e.g. Schmidt, Tippenhauer & Walach, 2001). A theoretical framework for this can be found, for example, *withinLucadou‘s Model of Pragmatic Information (Lucadou, 1995, 2001). Whether such a circumstance is responsible for our results can be only determined by the development and empirical testing of a model for Psi that accounts for such intrinsic properties

Talk about a get out clause for not being able to produce replicable effects!

 

[CFLarsen]

Yeah....but if they can't replicate it, how the heck are they going to develop a model that accounts for that?

 

[davidsmith73]

Ooh, I wouldn't bring up Occam in a psi discussion.

In this particular instance, I don't see why not. If the presentiment results can be accounted for by one of either PK, clairvoyance or precognition then its unnecessary to suppose its a combination.

 

[davidsmith73]

Now you are catching on to the issues we have been raising for some time. These are fairly basic and obvious issues when it comes to test protocol design. They bear thinking about BEFORE rushing forward to the test itself and subsequent publication...of questionable results.

I've always been aware of these flaws in the Sheldrake experiments. That's not to say that they can be corrected resulting in a tighter protocol using known callers, which is what we've been discussing.

Considering the original claims as provided, it is actually NOT unreasonable at all.

The "original claim" seems to be that the recipient knows, telepathically, when a caller known to the participant will call. Thus, an experiment using known callers is actually a good test of the "original claim". Why introduce all that noise from unknown callers?

And yes, it is certainly feasible to design a practical and representational protocol that uses a sample population of, say, 1000 random people from the phone book that are not reasonably known to the recipient. That allows quantification of the results as well.

Is the recipient going to be provided with a list of the names of all these callers?

If so, then you have an experiment basically identical with Sheldrakes except you are using far more unknown callers. That's fine, but it doesn't really make sense to have so many. You may as well just use 2 known and 2 unknown to make the stats simpler.

If not, then you are going to have a hell of time trying to estimate the likelyhood of the recipient getting the name of the caller correct by chance. Would you like to try?

The issue with the "limited set of known callers" protocol, while valid and interesting in itself,

Good. I'm glad you finally agree that its a valid protocol after all.

is that it is NOT testing the claim being made, even in a prototype mode. The original claim is a test of an actual "psi" capability - the ability to view/recognise another person at a distance without using normal means (in fact, the phone call is merely a synchronising mechanism, and not really vital in itself).
Instead, the limited protocol tests something else (primarily the ability to beat the odds).

An experiment using 4 known callers tests precisely the hypothesis you define above. Assuming a well controlled experiment, if 25% hit rate is expected by chance and we get 40% then we have positive result for "the ability to view/recognise another person at a distance without using normal means". And yes, this is also beating the odds! But beating the odds means that we have "the ability to view/recognise another person at a distance without using normal means".

Therefore any effect measured probably cannot be related to or translated to the original claim. Being good at the latter does not imply strength at the former.

Yes it does!! If we have controlled for "normal means" then beating the odds must mean that we have "the ability to view/recognise another person at a distance without using normal means"

 

[davidsmith73]

How will you ever falsify that theory?

I guess that the hypothesis predicts that changing the conditions of the observation will modulate the PK effect. If observation is not important then this won't happen. Thats just my initial thoughts. I certainly don't have a fleshed out theory for you!

How will you find out?

Systematically increase the number of observers of the data and look for changes in the effect or something along those lines. I don't think the experimental effect is robust enough for this yet.